Whole Genome Sequencing
Whole genome sequencing (WGS) determines the complete DNA sequence of an organism's genome in a single experiment. This guide is for molecular biologists, clinical researchers, and bioinformaticians who need a source bounded, practical framework to understand when to use WGS, how to execute a robust workflow, and what limits apply when interpreting results. WGS provides a base by base readout of an entire genome, making it the most comprehensive genetic assay available NCBI Bookshelf.
Unlike targeted sequencing, whole genome sequencing captures coding and noncoding regions, structural variants, and repetitive elements in one pass. The method has transformed fields from infectious disease outbreak tracking to cancer genomics and conservation biology. For example, researchers used WGS to characterize carbapenem resistant Klebsiella pneumoniae isolates in children Clinical and molecular characterizations of carbapenem resistant Klebsiella pneumoniae among children with bloodstream infections in Beijing, China. Another group applied whole genome data to resolve taxonomic confusion in Serratia species The current discordance on Serratia spp. taxonomical diagnosis using proteomics or genomic tools. These examples show the breadth of WGS applications, but success depends on careful planning and execution.
At a Glance
| Aspect | Key Detail |
|---|---|
| Definition | Complete sequencing of an organism's nuclear or organellar DNA |
| Typical read depth | 30x to 60x for human, 50x to 100x for bacteria |
| Library prep time | 1 to 2 days (short read), 3 to 7 days (long read) |
| Sequencing run time | 1 to 44 hours depending on platform and output |
| Main output files | FASTQ (raw reads), BAM/CRAM (aligned reads), VCF (variants) |
| Major cost components | Library construction, sequencing, compute for analysis |
| Common quality metrics | Q30 scores, mean coverage, duplication rate, insert size |
Core Concepts and Decision Points
Before starting a WGS project, you must decide whether the comprehensive nature of WGS justifies the cost and complexity compared to whole exome or targeted panels. For organisms with small genomes like bacteria, WGS is often the default because it provides full coverage for a lower per sample cost. For larger genomes like human, the main decision factor is the need for noncoding regions and structural variants.
A second decision point is read length versus accuracy. Short read platforms (e.g., Illumina) offer high accuracy (Q30 > 85 percent) but struggle with repetitive regions and large structural variants. Long read platforms (e.g., Oxford Nanopore, PacBio) resolve these features but have higher per base error rates that require polishing. Many pipelines now use hybrid assemblies combining both technologies.
The third major decision involves sequencing depth. Deeper coverage gives more confidence for rare variant detection but increases cost. For germline human WGS, 30x is a common target. For cancer samples or microbial genomes, depth may range from 50x to 100x. You can find detailed guidance on coverage planning from the EMBL EBI training materials EMBL EBI Training.
The choice of reference genome also matters. Use the most recent assembly for the species. For non model organisms, a reference may not exist, requiring de novo assembly. The Galaxy Training Network provides workflows for both resequencing and de novo projects Galaxy Training Network.
A Practical Workflow or Implementation Sequence
The following steps outline a standard WGS pipeline. Adapt the order and tools based on your organism, project scale, and available infrastructure.
Sample preparation and DNA extraction. Extract high molecular weight DNA with minimal degradation. Quantify using fluorometry and assess purity with spectrophotometry. Contaminants can inhibit library prep and reduce sequence yield.
Library construction. Fragment DNA to the desired size (typically 300 to 600 base pairs for short reads). End repair, A tail, and ligate adapters. For PCR free libraries, skip amplification to reduce bias. For low input samples, limited cycle PCR is acceptable.
Sequencing. Load the library onto a flow cell. Monitor run metrics including cluster density and Q scores. Discard runs where Q30 values fall below 80 percent for the majority of bases.
Raw data quality checks. Use FastQC or similar tools to examine per base quality, GC content, adapter contamination, and duplication rates. Trim low quality bases and remove adapters with cutadapt or Trimmomatic.
Alignment or assembly. For resequencing, align reads to the reference genome with BWA MEM or Minimap2. For de novo projects, assemble reads using SPAdes or MEGAHIT for short reads, or Flye and Canu for long reads.
Variant calling or genome annotation. Call single nucleotide variants and small indels using GATK HaplotypeCaller or DeepVariant. For structural variants, use tools like Manta or Sniffles. Annotate variants with SnpEff or ANNOVAR. For bacterial genomes, annotate with Prokka.
Quality filtering. Apply hard filters based on read depth, quality scores, and allele balance. Remove variants that fall in low complexity regions or homopolymer runs. The tradeoffs in filtering are well documented in non model organisms Navigating tradeoffs in variant filtering for population genetic and demographic inferences to inform management, conservation, and domestication in non model marine bivalves: a case study in scallop.
Interpretation and reporting. Compare variant frequencies to population databases. For clinical applications, assess pathogenicity using ACMG guidelines. For outbreak tracking, construct phylogenies using maximum likelihood or Bayesian methods.
Common Mistakes
Mistake 1: Insufficient coverage for the intended analysis. Low depth leads to high false negative rates, especially for heterozygous sites. Always calculate the required coverage before starting.
Mistake 2: Ignoring sample contamination. Mixed samples produce spurious variant calls. Include negative controls and check for cross contamination using tools like VerifyBamID.
Mistake 3: Overly aggressive or overly lenient variant filtering. As shown in the scallop study, filtering thresholds dramatically affect downstream demographic inferences. Validate your filter set with known positive and negative controls.
Mistake 4: Using outdated or mismatched reference genomes. Alignment rates drop and variant call quality suffers when the reference is not current. For human, use GRCh38 with patches. For bacteria, use the most complete assembly.
Mistake 5: Neglecting metadata. Without accurate sample identifiers, extraction dates, or library barcodes, results cannot be traced. Maintain a detailed lab notebook and use barcodes consistently. The NCBI Sequence Read Archive requires complete metadata for public submission NCBI Sequence Read Archive.
Limits and Interpretation
Whole genome sequencing cannot detect all genetic variation. It fails in highly repetitive regions, such as centromeres and telomeres, where short reads cannot map uniquely. It also cannot distinguish between identical DNA segments located in different parts of the genome. Epigenetic modifications like DNA methylation are not directly captured unless bisulfite sequencing is used.
Interpretation is further limited by gaps in reference genomes. Even the human reference has unfinished regions. For non model organisms, the reference itself may have many errors. The study on Cypripedium orchids highlighted that chloroplast genome datasets, while valuable, come from limited populations and may not represent global diversity A population scale chloroplast genome dataset for the endangered orchids Cypripedium taibaiense and C. flavum.
Variant validation remains essential. No sequencing platform achieves 100 percent accuracy. Confirmation using Sanger sequencing or orthogonal genotyping is recommended for high stakes findings. In clinical settings, the genetic architecture of complex traits like lung cancer involves both common and rare variants, and WGS alone cannot establish causality without functional studies Genetic architecture of lung cancer revealed by common and rare variant analyses across population scale biobanks.
Frequently Asked Questions
What is the difference between whole genome sequencing and whole exome sequencing? Whole genome sequencing reads all DNA in the genome, including introns and intergenic regions. Whole exome sequencing targets only the protein coding exons, which account for about 1 to 2 percent of the human genome. WGS provides more comprehensive data at a higher cost.
How long does a whole genome sequencing run take? Turnaround time depends on the platform. Illumina runs typically take 1 to 3 days for human scale coverage. Nanopore flow cells can produce data in real time, with a full genome in 1 to 2 days. Library preparation adds 1 to 2 days.
Can whole genome sequencing detect structural variants? Yes, but detection sensitivity varies. Large deletions, duplications, and inversions are recoverable with both short and long reads. Balanced rearrangements like translocations are harder to identify with short reads. Long read sequencing significantly improves structural variant detection.
What is the typical cost for whole genome sequencing in 2025? For human genomes, reagent and sequencing costs range from $500 to $1,500 per sample at commercial scale. Bioinformatics analysis adds additional compute and personnel costs. For bacterial genomes, costs are much lower, typically $50 to $150 per sample.
References and Further Reading
- NCBI Bookshelf , Authoritative technical references on sequencing technologies and genomics.
- EMBL EBI Training , Official training resources for biological data analysis, including WGS workflows.
- Galaxy Training Network , Open bioinformatics training materials with hands on WGS tutorials.
- Bioconductor , Open software and documentation for genomic data analysis in R.
- NCBI Sequence Read Archive , Public repository for high throughput sequencing data and metadata.
- Clinical and molecular characterizations of carbapenem resistant Klebsiella pneumoniae among children with bloodstream infections in Beijing, China , Example WGS application in infectious disease.
- Navigating tradeoffs in variant filtering for population genetic and demographic inferences to inform management, conservation, and domestication in non model marine bivalves: a case study in scallop , Practical guidance on filtering thresholds.
- The current discordance on Serratia spp. taxonomical diagnosis using proteomics or genomic tools , WGS for bacterial taxonomy.
- Tracing and characterization of a family outbreak of subtype B2 botulism linked to homemade pickled eggs in Jinan, China , WGS in outbreak investigation.
- Genetic architecture of lung cancer revealed by common and rare variant analyses across population scale biobanks , WGS in cancer genetics.